Skip to content

abhinavnandwani/abhinavnandwani.github.io

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

97 Commits
 
 
 
 
 
 
 
 

Repository files navigation

👋 About Me

I am a undergradute majoring in computer engineering at uw-madison. My interests lie in energy efficent neural network inference ASIC design.

📚 Education

  • University of Wisconsin-Madison, College of Engineering
    Bachelor of Science, Computer Engineering (Machine Learning Specialization), Computer Science Minor
    GPA: 3.86/4.0
    Expected Graduation: 2026

🔬 Projects and Experiences

ARM-Based AI Inference Accelerator for Meta Llama 2 Repo

  • Developed an ASIC on the AMD Xilinx platform for a scaled-down 50M Llama 2 model using a systolic array optimized for INT8 quantization in SystemVerilog.
  • Implemented the Llama 2 architecture from scratch in C, leveraging an ARM Cortex microprocessor.
  • This architecture increased power efficiency by offloading intensive matrix multiplications, allowing the ARM chip to focus on general system operations.

Pipelined RISC CPU Repo

  • Designed, in collaboration, a 5-stage pipelined MIPS CPU in Verilog with control, data hazards prevention and exception handling.
  • Integrated a two-way set-associative cache with four-banked memory and optimizations like bypassing, forwarding, and branch prediction to minimize pipeline stalls.

Autonomous Robot Chip Synthesis Repo

  • Worked in a team of four, under Eric Hoffman’s guidance, to design an autonomous robot capable of executing the entire Knight's Tour on a chessboard, entirely at the RTL level.
  • Employed a PID controller with PWM for precise movement and positioning. Utilized UART and SPI interfaces for effective communication between the robot’s components and the microprocessor.
  • Simulated and Synthesized on Intel Quartus and Synopsys using a 32nm library. Applied PVT and low-level constraints, optimizing the design and meeting max and min delay slack with clock uncertainty.

📘 Advanced Courses I've Taken

  • ECE 551: Digital Design and Synthesis
  • ECE 552: Computer Architecture
  • ECE 532: Matrix Methods in Machine Learning
  • CS 577: Introduction to Algorithms
  • ECE 352: Digital System Fundamentals
  • ECE 354 : Machine Organization and Programming (The C language)

📈 Skills

  • Languages: SystemVerilog, Python, C++, C
  • Tools: Synopsys Design Vision, ModelSim / Questa , PyTorch, Tensorflow
  • Core Competencies: ASIC Design, Digital System Design, CPU Architecture, Machine Learning Algorithms, Neural Network Inference Optimization

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published